Closed Klazkin closed 3 months ago
Refactoring the MCTS class so the code is less horrible and its easier to implement new features and to integrate it with new models.
Time Estimate: 5 hours 0 minutes Time spent: 6 hours 30 minutes
5 hours 0 minutes
6 hours 30 minutes
Monte Carlo Tree Search: A Review of Recent Modifications and Applications https://arxiv.org/pdf/2103.04931v4.pdf A Survey of Monte Carlo Tree Search Methods https://www.researchgate.net/publication/235985858_A_Survey_of_Monte_Carlo_Tree_Search_Methods Learning Policies from Self-Play with Policy Gradients and MCTS Value Estimates https://arxiv.org/pdf/1905.05809.pdf
Refactoring the MCTS class so the code is less horrible and its easier to implement new features and to integrate it with new models.
The goal
Time tracking
Time Estimate:
5 hours 0 minutesTime spent:6 hours 30 minutesResources
Monte Carlo Tree Search: A Review of Recent Modifications and Applications https://arxiv.org/pdf/2103.04931v4.pdf A Survey of Monte Carlo Tree Search Methods https://www.researchgate.net/publication/235985858_A_Survey_of_Monte_Carlo_Tree_Search_Methods Learning Policies from Self-Play with Policy Gradients and MCTS Value Estimates https://arxiv.org/pdf/1905.05809.pdf